The researchers said they found a strong case for taking away some responsibility for ward rounding from busy acute care nurses.
“This study emphasises the importance of early warning scores to identify deterioration”
They compared two models of “rapid response systems” to identify patients at risk of clinical deterioration at a hospital in Florida.
The first was described as a “reactive” approach, in which ward nurses were required to monitor patients and manually call for help from a rapid response team if they took a turn for the worse.
The researchers said the rapid response teams varied in composition, but almost always included a registered nurse with intensive care unit experience. The team was “structurally empowered to intervene to facilitate timely diagnosis and management of the deteriorating patient”.
Under the reactive model, the team was manually “activated” by paging the designated rapid response nurse who would be at the patient’s bedside within five minutes to assess them and call for additional clinicians on a case-by-case basis.
In addition to responding to rapid response team activations, the designated nurse also had other patient care responsibilities in the ICU.
The second model was a “proactive” approach and involved the use of an electronic early warning score system called the Rothman Index. The index combined the patient’s vital signs, laboratory test results and nurse assessment data to track their condition over time, updating up to once per hour.
“They could focus exclusively on the surveillance function and were not likely subject to the same degree of distractions”
The information was converted into colour-coded graphs and results could be viewed for individual patients or for multiple patients at once.
In the study, published in the International Journal of Nursing Studies, the graphs were monitored by an intensive care nurse at the start of their 12-hour shift who would use the data to identify patients at risk of deterioration.
The nurse would then generate a list of patients to be included in routine surveillance rounds from the rapid response team. Under this system, the rapid response team nurses did not carry any additional duties in the intensive care unit.
The study analysed 12,148 patients over two separate six months periods. The first sample was collected between October 2010 and March 2011, when staff were using the reactive model, and the second between October 2011 and March 2012, when the proactive model was in operation.
The researchers found that unplanned ICU transfers were 1.4 times more likely to occur using the reactive model compared to proactive. Or putting it differently, there was a 40% decrease in ICU transfers under the proactive system.
The authors noted that unplanned escalations in care such as transfers to an ICU “place patients at a greater risk for hospital mortality, greater severity of illness, and longer hospital stays”.
The implementation of the Rothman Index resulted in 1,440 out of the 6,273 patients reviewed being referred for proactive surveillance rounds by the rapid response team – a 312% increase compared to the reactive model.
The authors of the study said this finding reflected “significant improvement” in the “afferent limb” of the process, in which the risk of clinical deterioration was detected, and the rapid response team was called.
They said: “Our findings suggest that afferent limb failures in this model can be reduced by shifting some of the responsibility for event detection and trigger response activation from nurses on acute care units to nurses in the ICU with special training, access to automated early warning scores, and dedicated time to conduct proactive surveillance rounds.
“The enhanced efficiency of automated early warning scores presented in color-coded graphical displays made it possible for one nurse to provide timely estimation of risk for clinical deterioration across multiple patients from one remote ICU location,” stated the researchers.
They added: “Moreover, because the [proactive] model nurses did not carry additional patient care assignments, they could focus exclusively on the surveillance function and were not likely subject to the same degree of distractions, interruptions, and cognitive shifting inherent to the role of acute care nurses.”
The authors added that the increase in rapid response team activations could also be related to the “empowerment” of the ICU nurses to initiate surveillance rounds based on objective data from the Rothman Index rather than waiting for an invitation from acute staff.
“This shift in responsibility for the response trigger to the ICU nurses effectively bypassed trigger choice options for acute care staff,” they said.
“The risk for delay or avoidance of a response trigger based on implicit biases and other contextual influences related to hierarchical organisational cultures was removed,” noted the authors.
The study was the first to document a positive effect of proactive surveillance guided by automated early warning score data on unplanned ICU admissions.
However, the researchers admitted the study had limitations, including the fact that it was carried out in a single hospital.
Reacting to the findings, Professor Peter Griffiths, chair of health services research at the University of Southampton, told Nursing Times: “I think this study emphasises the importance of early warning scores to identify deterioration and, more importantly, the need to act on the scores when risk is identified.
“This model may not work for every context but the idea of rapid response teams actively rounding to seek out patients at risk, based on an early warning score, is worth exploring further,” he said.
However, Professor Griffiths said he would have preferred for the study to have included information on the costs and also on the benefits to patients in terms of reduced mortality and adverse events beyond the reduction in unplanned ICU admission.